Deciphering the genetic architecture of variation

in the immune response to Mycobacterium

tuberculosis infection
Luis B. Barreiroa,b,1,2, Ludovic Tailleuxc,1,2, Athma A. Paia, Brigitte Gicquelc, John C. Marionia,d, and Yoav Gilada,2
a

Department of Human Genetics, University of Chicago, Chicago, IL 60637; bSainte-Justine Hospital Research Centre, Department of Pediatrics, Faculty of
Medicine, University of Montreal, Montreal H3T 1C5, QC, Canada; cUnit of Mycobacterial Genetics, Institut Pasteur, Paris 75015, France; and dEuropean
Bioinformatics Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge CB10 1SD, United Kingdom
Edited by Barry R. Bloom, Harvard School of Public Health, Boston, MA, and approved December 15, 2011 (received for review September 24, 2011)

Tuberculosis (TB) is a major public health problem. One-third of the

worlds population is estimated to be infected with Mycobacterium tuberculosis (MTB), the etiological agent causing TB, and active disease kills nearly 2 million individuals worldwide every year.
Several lines of evidence indicate that interindividual variation in
susceptibility to TB has a heritable component, yet we still know
little about the underlying genetic architecture. To address this,
we performed a genome-wide mapping study of loci that are associated with functional variation in immune response to MTB.
Specically, we characterized transcript and protein expression
levels and mapped expression quantitative trait loci (eQTL) in primary dendritic cells (DCs) from 65 individuals, before and after
infection with MTB. We found 198 response eQTL, namely loci that
were associated with variation in gene expression levels in either
untreated or MTB-infected DCs, but not both. These response eQTL
are associated with natural regulatory variation that likely affects
(directly or indirectly) host interaction with MTB. Indeed, when we
integrated our data with results from a genome-wide association
study (GWAS) for pulmonary TB, we found that the response eQTL
were more likely to be genetically associated with the disease. We
thus identied a number of candidate loci, including the MAPK
phosphatase DUSP14 in particular, that are promising susceptibility genes to pulmonary TB.

uberculosis (TB) is among the oldest diseases recorded to

affect humans. Skeletal remains show that prehistoric
humans (7000 BC) had already suffered from TB (1) and tubercular decay has been found in the spines of mummies from
3000 to 2400 BC (2). Today, despite considerable efforts to ght
the disease, TB remains a major public health problem, with 1.7
million deaths occurring annually worldwide and up to one-third
of the global population estimated to be carrying latent Mycobacterium tuberculosis (MTB) infection (3). A striking feature of
TB is that only 10% of individuals infected with MTB develop
the disease (4, 5). Although a signicant proportion of interindividual variation in susceptibility to TB can be attributed to
environmental factors such as malnutrition or poor hygienic
conditions, a substantial portion is thought to be due to host
genetic factors (68). The strongest evidence for this probably
comes from twin studies showing that the rate of TB in monozygotic twins is more than twice that observed among dizygotic
twins (8). In addition, studies on Mendelian susceptibility to
mycobacterial disease (MSMD) have identied multiple rare
single-gene mutations linked with susceptibility to mycobacteria
(7, 9, 10). Although studies on MSMD have played a key role in
identifying important pathways involved in protective immunity
against TB, such as the IL12/23- and IFN- pathways (7, 914),
the mutations identied are too rare to have a signicant impact
on the overall variation in susceptibility to TB in the wider
population.
The quest of the genetic determinants of susceptibility to TB
at the population level has so far been primarily driven by case
control studies of candidate genes. Studies based on such approaches identied over 20 genes associated with susceptibility
12041209 | PNAS | January 24, 2012 | vol. 109 | no. 4

to TB (reviewed in refs. 6, 7). However, limited samples sizes and

difculties in validating the majority of these ndings across
populations raised doubts about many of these associations (15).
Recently, Thye et al. (16), reported the rst genome-wide association study (GWAS) for host susceptibility to pulmonary TB
(using 2,237 cases and 3,122 controls from West Africa). Despite
the relatively large sample size, only a single locus in a gene desert
region was associated with a signicant genome-wide modest
effect size (odds ratio = 1.2). Thus, to date, little is known about
the underlying genetic determinants or mechanisms contributing
to differences in susceptibility to TB at the population level.
In the context of an infectious disease such as TB, the most
important molecular networks affecting disease susceptibility are
probably those involved in mechanisms of immune defense. We
thus reasoned that genetic variants that are associated with
variation in immune response to MTB infection would be highly
promising genetic candidates for susceptibility to TB. To identify
such variants, we characterized genome-wide gene expression
levels and mapped expression quantitative trait loci (eQTL) in
untreated and MTB-infected monocyte-derived dendritic cells
(DCs), from a panel of 65 healthy individuals of European descent. We chose to work with DCs because they play a central
role in bridging innate and adaptive immunity. Upon MTB infection, DCs migrate from the lungs to the draining lymph nodes
where they present antigens to naive T cells, orchestrating antimycobacterial immunity and ultimately determining disease
outcome (17). We thus provide a comprehensive view of regulatory mechanisms underlying variation in immune response to
MTB infection.
Results
Transcriptional Response of DCs to MTB Infection. We infected DCs

from 65 healthy individuals with a virulent strain of MTB. Following infection, we extracted RNA from the untreated and
infected DCs at the same time, 18 h after the infection. We then
characterized genome-wide gene expression proles in all samples using the Illumina HT-12 expression arrays. After excluding
data from poorly annotated array probes and from genes that
were classied as not expressed, we normalized the expression
data for the remaining 12,958 genes (details in Materials and

Author contributions: L.B.B., L.T., and Y.G. designed research; L.B.B. and L.T. performed
research; L.T., A.A.P., B.G., J.C.M., and Y.G. contributed new reagents/analytic tools; L.B.B.
analyzed data; and L.B.B. and Y.G. wrote the paper.
The authors declare no conict of interest.
This article is a PNAS Direct Submission.
Freely available online through the PNAS open access option.
Data deposition: The gene expression and genotype data reported in this paper have
been deposited in the Gene Expression Omnibus (GEO) database (accession nos. GSE34151
and GSE34588, respectively).
1

L.B.B. and L.T. contributed equally to this work.

To whom correspondence may be addressed. E-mail: gilad@uchicago.edu, luis.barreiro@

umontreal.ca, or tailleux@pasteur.fr.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.

1073/pnas.1115761109/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1115761109

Methods). We then analyzed the gene expression data using

a linear model with a xed effect for the treatment (infection
with MTB; Materials and Methods). Using a moderated t test
(18), we classied 2,948 and 4,055 genes as up- or down-regulated postinfection, respectively (P < 106, false discovery rate
(FDR) < 104). The difference in expression level between the
untreated and infected DCs was greater than 0.5-fold for 3,040
(43%) of these genes (Fig. 1A and Dataset S1).
Changes in the transcriptome following MTB infection are
expected to reect the transition of DCs from antigen-capturing
cells to potent antigen-presenting cells and T-cell activators (19,
20). Consistent with this notion, we found that the maturation of
DCs after MTB infection was accompanied by the strong upregulation of genes involved in immune responses (FDR < 1026),
including cytokine signaling, T-cell activation, and antigen presentation (FDR < 0.006; Fig. 1B and Dataset S2). Conversely,
down-regulated genes were signicantly enriched for genes involved in metabolic pathways (FDR < 106) and in the reorganization of the cytoplasmic membrane (FDR < 106, Fig. 1B
and Dataset S2). This observation probably reects a reduced
capacity of mature DCs to endocytose/phagocytose and the
expected reduction in trafcking between the mycobacterial
phagosome and the host cell recycling and biosynthetic pathways (20, 21).

Mapping Cytokine and Chemokine Secretion Levels. In addition to

characterizing transcriptional proles following MTB infection,
we measured the levels of 19 cytokines in the supernatants of
untreated and infected DCs (Fig. 3 and Fig. S2; Materials and
Methods for details). With the exception of IL-17, the protein
measurements of all cytokines/chemokines showed a signicant
correlation with estimates of the corresponding transcript expression levels (P < 0.01; median Spearman correlations = 0.6;
Fig. 3A). As expected, the expression levels of all tested proteins
were signicantly elevated after MTB infection (Wilcoxon
signed-rank test, FDR < 0.01, Fig. S2), and we observed strong
induction of all of the cytokines presently known to play a critical

GENETICS

Identication of Immune Response eQTL. We next aimed to identify

eQTL in both infected and noninfected DCs. To do so, we
genotyped each individual at 970,287 single nucleotide polymorphisms (SNPs) using the Illumina Omni1-Quad BeadChip.
Genotypes for 873,973 SNPs passed our quality checks (Materials
and Methods). We used a linear regression model to test for an
association between the expression levels of 11,996 autosomal
genes and genotype at all SNPs with a minor allele frequency
greater than 10% (SI Materials and Methods includes specic
details on data processing and modeling). We mapped eQTL
separately in infected and noninfected DCs. In agreement with
previous studies (e.g., refs. 22, 23), we found that most SNPs
strongly associated with gene expression levels lie near the corresponding gene (putatively acting in cis; Fig. S1). Thus, we focused our analysis on putative cis-regulatory variants dened as
SNPs located in a 200-kb window centered on a genes transcription starting site (TSS). At an FDR of 1% (at least one SNP
with P < 1.4 105; more details in Materials and Methods) we
found 720 and 756 genes with cis-eQTL in the infected and noninfected DCs, respectively (Fig. 2A and Dataset S3). Interestingly,

genes whose expression levels were altered following MTB infection were 1.6 times more likely to be associated with cis-eQTL
than the genome-wide average (14% compared with the 9% that
are expected by chance alone; P = 1.6 1016; Fig. 2B). Our
results thus suggest that genes involved in immune response to
MTB have increased levels of functional diversity in the population. Such functional variation is likely to underlie interindividual
differences in susceptibility to infectious diseases in general and
MTB infection in particular.
As expected, there is a large overlap of eQTL identied independently in the noninfected and infected DCs (Pearson
correlation of eQTL association P values in the two classes of
DCs is 0.68; P < 1015). In the context of susceptibility to TB,
however, the most interesting eQTL are arguably those with
a different effect on gene expression levels before and after infection with MTB, which we term response eQTL. Response
eQTL likely interact with MTB (directly or indirectly) and thus
may account for interindividual variation in immune response to
MTB infection. We classied response eQTL by using highly
conservative criteria to minimize the probability of false positives
(e.g., when true eQTL in both untreated and infected DCs are
only classied as such in one class because of incomplete power).
Specically, we dened response eQTL when we found strong
evidence for a cis-eQTL for a gene in either untreated or
infected DCs at an FDR of 1% (P < 1.4 105) and no statistical
evidence supporting a cis-eQTL for the same gene (in the entire
tested 200-kb region) in the other condition at a very relaxed
FDR threshold of 50% (more details in Materials and Methods).
Using this approach, we identied 198 genes with strong evidence of being associated with at least one response eQTL (102
and 96 eQTL in untreated and infected DCs, respectively; Fig. 2
A, C, and D and Dataset S3).

Fig. 1. Functional characterization of immune responses to MTB infection. (A) Volcano plot showing differentially expressed genes after infection of DCs
with MTB for 18 h. The negative log10 transformed P values test the null hypothesis of no difference in expression levels between untreated and infected DCs
(y axis) and are plotted against the average log2 fold changes in expression (x axis). Data for genes that were not classied as differentially expressed are
plotted in black. In gray and blue, we plotted data for genes that are differentially expressed after infection with MTB (P value <7.7 107; Bonferroni
corrected P value <0.01) with an absolute log2 fold change (|FC|) less than or equal to 0.5 or greater than 0.5, respectively. (B) Gene ontology (GO) enrichment
analysis for genes that were classied as up- (red) or down- (blue) regulated following infection of DCs with MTB. Only signicant enrichments at an FDR <1%
are plotted (complete results in Dataset S2).

Barreiro et al.

PNAS | January 24, 2012 | vol. 109 | no. 4 | 1205

Fig. 2. Deciphering the genetic basis of interindividual variation in immune response to MTB infection. (A) Plot contrasting the evidence for cis-eQTL in the untreated and
infected DCs. For every gene we plotted the additive model
P values (log10 transformed) for the most strongly associated
cis-SNP (dened as SNPs located in 200-kb window centered on
the TSS of a proximal gene) with gene expression levels in the
untreated (x axis) or infected (y axis) DCs, respectively. The red
dashed lines specify the P values corresponding to an FDR of
1%. The blue dashed lines specify the second, more relaxed,
cutoff (50% FDR) used to condently classify response eQTL.
Only genes with strong evidence of a cis-eQTL in at least one of
the conditions (FDR of 1%) are plotted. (B) Proportion of ciseQTL (y axis) observed among all tested genes and among
genes that were classied as differentially expressed (DEG)
following infection with MTB. (C) Example of a response eQTL
found only in the untreated samples. (D) Example of a response eQTL found only in the infected samples.

role in protective immunity against TB (2426) (Fig. 3B), including TNF- (460-fold induction), IFN- (24-fold induction),
and IL-12 (8-fold induction).
Using the protein secretion data we also looked for protein
QTL (pQTL), namely loci that are associated with cytokines/
chemokines secretion levels measured after infection of the

DCs. We did not nd any signicant association, either genomewide or when we restricted the analysis to SNPs within 200 kb of
the TSSs of the genes encoding the measured proteins. However,
when we only considered the SNPs that were previously identied as cis-eQTL (in the analysis of transcript levels above) we
found a clear enrichment in the association of genotypes with

Fig. 3. Association between cis-eQTL and protein

secretion levels. (A) Boxplot of Spearman correlations between mRNA and protein expression levels.
(B) Examples of data from TNF-, IFN-, and IL-12
(from Left to Right). Protein level measurements
from untreated (green) and infected (red) DCs from
each of the 65 individuals are plotted. Results for
the remaining proteins can be found in Fig. S2.
P values were obtained using a nonparametric
Wilcoxon test, which takes into account the paired
nature of our data. (C) Manhattan plot showing the
negative log10 transformed P values (y axis) for the
association between all SNPs classied as cis-eQTL
and the secretion levels of IL-1Ra measured in the
supernatant of infected DCs. (D) Correlation between genotypes at rs11960575 and the relative
secretion levels of IL-1Ra. In addition to the association between rs11960575 and IL-1Ra secretion
levels, we also found a signicant association
(Bonferroni P < 1.7 106) between rs854100 and
the secretion levels of IL-15, although the secretion
levels of IL-15 after infection were very low (Figs. S2
and S3).

1206 | www.pnas.org/cgi/doi/10.1073/pnas.1115761109

Barreiro et al.

protein levels (Fig. S3). This observation is tentative and requires

further validation and replication, but it suggests that the secretion of a subset of these cytokines is regulated in trans by
proteins whose expression levels are, in turn, regulated in cis by
the eQTL we found. In particular, in two cases (of the 19 tested),
the association was signicant even after the conservative Bonferroni correction for multiple testing. Specically, rs854100 and
rs11960575 are associated with the secretion levels of IL15 and
IL-1Ra, respectively (Bonferroni P < 1.7 106; Fig. 3 C and D
and Fig. S3). The genetic variation at rs11960575 may be of
particular interest: This locus was classied as a cis-eQTL for the
gene brillin 2 (FBN2; Pnoninfected DCs = 8.3 109; PMTB-infected
3
DCs = 2 10 ), and a trans-pQTL for IL-1Ra (Bonferroni P <
6
1.7 10 ; Fig. 3 C and D). This observation may reect a regulatory interaction between FBN2 and IL-1Ra. Consistent with
this inference, the members of the brillin gene family are
known to be involved in the storage and activation of TGF- (27,
28), which in turn has been shown to regulate the production of
IL-1Ra (29, 30). Because genetic variation in IL1Ra has previously been associated with susceptibility to TB (31, 32),
rs11960575 (or another SNP in linkage disequilibrium with it)
represents a strong candidate locus for impacting susceptibility
to TB.
Immune Response eQTL Are Enriched for Susceptibility Genes for TB.

Discussion
We studied variation in the regulatory response to MTB infection of DCs. We chose to focus on DCs because they have
been shown to play an essential, nonredundant role in protective

GENETICS

The immune response eQTL as a class are strong candidates for

affecting susceptibility to TB. To test this possibility, we integrated our data with results from the only GWAS for host
susceptibility to pulmonary TB published to date, which included
2,237 cases and 3,122 controls from Ghana and Gambia (16). In
the original GWAS analysis, only a single locus, in a gene desert
region on chromosome 18q11, reached genome-wide signicance
(16). When we focused on the response eQTL identied in this
study, however, we found that these loci were more likely to be
associated with TB than expected by chance (Fig. 4). Specically,
we observed more than 1.4-fold enrichment of SNPs with a nominal GWAS P value <0.05 among response eQTL, compared with
genome-wide expectations (Fig. 4B, 2 test, P = 0.012). Impor-

tantly, we did not observe a similar enrichment when we considered eQTL that are common to untreated and infected DCs (Fig.
4B; this observation serves as a control for a possible bias in the
power to detect an association in the GWAS when only eQTL
are considered).
We note that our observations are robust to the particular
method used to identify response eQTL. Indeed, an alternative
approach used to identify cis-interactions (i.e., response-eQTL)
is to consider variation in the magnitude of the shifts in gene
expression after a treatment, in our case MTB infection, as the
quantitative trait to be mapped (33, 34). Supporting our previous
ndings, MTB-response eQTL identied using this approach
(Dataset S3) were also signicantly enriched for GWAS P values
<0.05 (1.8-fold enrichment, 2 test, P = 0.01; Fig. S4).
Taken together, our results indicate that response eQTL are
likely enriched for susceptibility alleles for TB. In particular, our
data strongly support the notion that dual-specicity phosphatase 14 (DUSP14) is a new susceptibility gene for TB. Indeed, we
found strong evidence that rs712039 is an immune response
eQTL associated with variation in the expression levels of
DUSP14 exclusively in noninfected DCs (Pnoninfected DCs = 9.6
106 (FDR = 6 103); PMTB-infected DCs = 0.09 (FDR = 0.9);
Fig. 5A). In addition, in our focused GWAS analysis, which was
restricted to the set of SNPs we classied as immune response
eQTL (as the set of most likely candidates), the strength of the
association between rs712039 and pulmonary TB is signicant
even after a conservative Bonferroni correction (Fig. 4D). This
SNP was not discussed in the original GWAS paper, probably
because it was not signicant at a genome-wide threshold, yet it
shows one of the strongest genetic associations with pulmonary
TB, in both the Ghana cohort (P = 9.8 104) and the Gambia
replication cohort (P = 2.3 103, combined P = 3.3 106;
Table S1).

Fig. 4. MTB-response eQTL are strong candidates to impact susceptibility to pulmonary TB. (A) The median GWAS P value for an expanding window of genes
is plotted. We used the GWAS P values obtained when combining the Ghana and Gambia cohorts (16). Genes are ordered by the strength of evidence
supporting an association with an eQTL only in the untreated (Left) or the infected (Right) DCs, respectively. To avoid positional biases, we restricted our
analyses to the set of cis-SNPs that was tested in our study (i.e., SNPs located in 200-kb window centered on the TSS of proximal genes). (B) Histogram of the
proportion of GWAS SNPs with nominal P values <0.05 among all GWAS SNPs (gray), among SNPs that were classied as eQTL in both untreated and infected
DCs (blue), and among response eQTL (red). (C) Manhattan plot showing the negative log10 transformed P values (y axis) for the association between the
response eQTL identied in this study and susceptibility to pulmonary TB. The dashed line corresponds to the genome-wide signicance cutoff after
a conservative Bonferroni correction.

Barreiro et al.

PNAS | January 24, 2012 | vol. 109 | no. 4 | 1207

immunity to TB (35). DCs exhibit the unique ability to migrate to

secondary lymphoid organs where they present mycobacteriaderived antigens to naive T cells (17). In the absence of DCs, the
response induced by CD4+ T cells is impaired and bacterial load
uncontrolled (3638). Accordingly, individuals with a deciency
of DCs, monocytes, and B and natural killer (NK) cells (DCML
deciency) show increased susceptibility to poorly virulent
strains of Mycobaterium spp (39). Similarly, mutations in the IFN
regulatory factor 8, which have recently been shown to be

Fig. 5. Genes with response eQTL likely to impact susceptibility to pulmonary TB. (A) Response eQTL identied for DUSP14. (B) Response eQTL
identied for RIPK2. (C) Response eQTL identied for ATP6V0A2.

1208 | www.pnas.org/cgi/doi/10.1073/pnas.1115761109

associated with a selective loss of CD11c+ CD1c+ myeloid DCs,

result in a remarkable increased susceptibility to disseminated
bacillus CalmetteGurin-osis, arguing for a specic role for
DCs in resistance to mycobacteria (40).
It is well established that macrophages (M) also play a central role in TB pathogenesis (2526). They are one of the primary cell targets of MTB, which have developed different
strategies to survive and to multiply inside the M phagosome
such as prevention of phagosome acidication (41) and inhibition of phagolysosomal fusion (42). In addition, macrophages
play an important role in determining the outcome of infection
by orchestrating the formation of granulomas and by killing the
bacillus upon IFN- activation. Thus, it would be of great interest
to also study the transcriptional response and map response
eQTL to MTB infection of M. Extending this approach to
additional relevant cell types, such as M, should greatly increase our understanding of the genetic architecture of TB response, with the ultimate aim of deciphering the genetic factors
controlling human susceptibility to this ancient disease.
Our results indicate that variation in the regulatory immune
response to MTB infection of DCs is often associated with genetic
factors. Indeed, at least 14% of differentially expressed genes
following infection with MTB are associated with an eQTL (Fig.
2B). This observation strongly supports the hypothesis that genetic susceptibility to TB is likely to be accounted for by the
combination of many genetic variants with small effects. Under
this model, classical GWAS approaches alone may be insufcient
for fully characterizing the multitude of genetic associations with
TB susceptibility (a challenge that is shared with a large number
of other complex phenotypes, including many other infectious
diseases). Our ndings indicate that in vitro experimental studies
of the eQTL response to infection represent a powerful approach
that can help overcome this challenge, as well as point to the likely
regulatory mechanisms that affect disease susceptibility.
Our data revealed DUSP14 as a susceptibility gene to pulmonary TB. The DUSP14 gene is a member of a large family of
phosphatases that specically dephosporylate threonine and tyrosine residues on mitogen-activated protein kinases (MAPKs),
rendering them inactive (43). These phosphatases therefore ultimately control the levels of proinammatory cytokines released
after infection (43). Accordingly, and further supporting a potential role of DUSP14 in susceptibility to TB, we found that the response eQTL we identied for DUSP14, rs712039, is also associated
with the secretion levels of TNF- (P = 0.02) and IFN- (P = 0.03).
These proteins are probably the two most important cytokines
known to be involved in immune protection against TB. These genes
participate in the formation and maintenance of the granulomes
(25) and activate the bactericidal activity of phagocytes (26).
Other novel strong candidates to affect susceptibility to pulmonary TB include ATP6V0A2 and RIPK2. Both genes are associated with response eQTL in MTB-infected DCs (Fig. 5 B and
C), and the same genetic variants are nominally associated with
pulmonary TB in the GWAS data (Table S1). RIPK2 encodes an
adapter protein of the NOD2-dependent signaling pathway, which
is a key pathway involved in the regulation of host responses to
MTB infection (4446). ATP6V0A2 encodes a subunit of the vesicular proton-ATPase (v-ATPase) (47). Interestingly, one major
virulence feature of the tubercle bacillus is its ability to parasitize
macrophages through the exclusion of v-ATPase from phagosomes, preventing them from maturing (41, 48). Interindividual
variation in the levels of ATP6V0A2 is therefore likely to impact
the ability of the bacillus to grow and survive inside phagocytic
cells, ultimately impacting susceptibility to TB.
It is important to note that our approach does not allow us to
distinguish between response eQTL that are specic to MTB
infection and those shared with other infectious diseases. Because we show that the response eQTL we identied are enriched for genetic variants associated with susceptibility to TB,
it is reasonable that at least a subset of these eQTL have specic
roles in the immune response to MTB infection. We expect that
our ndings will stimulate future work aimed at characterizing
Barreiro et al.

Materials and Methods

Details of the experimental and statistical procedures can be found in SI
Materials and Methods. Blood samples from 68 healthy donors were
obtained from Research Blood Components. All samples were collected
with informed consent under the companys independent ethics committee approval. All individuals recruited in this study were healthy Caucasian
males between the ages of 21 and 55 y old. Genotyping of all individuals
was performed using the Illuminas Omni1-Quad BeadChip array. After
a series of quality checks (SI Materials and Methods), genotyping data
from 65 samples were kept for downstream analyses. Blood mononuclear
cells from each donor were isolated by Ficoll-Paque centrifugation and
blood monocytes were puried from peripheral blood mononuclear cells
(PBMCs) by positive selection with magnetic CD14 MicroBeads (Miltenyi
Biotec). Monocytes were then derived into DCs as previously described (20)
and subsequently infected with MTB for 18 h at a multiplicity of infection
of 1-to-1. Genome-wide gene expression proles of untreated and infected

1. Hershkovitz I, et al. (2008) Detection and molecular characterization of 9,000-year-old

Mycobacterium tuberculosis from a Neolithic settlement in the Eastern Mediterranean. PLoS ONE 3:e3426.
2. Zink AR, et al. (2003) Characterization of Mycobacterium tuberculosis complex DNAs
from Egyptian mummies by spoligotyping. J Clin Microbiol 41:359367.
3. World Health Organization (2009) WHO Report 2009: Global Tuberculosis Control:
Epidemiology, Strategy, and Financing (WHO, Geneva).
4. North RJ, Jung YJ (2004) Immunity to tuberculosis. Annu Rev Immunol 22:599623.
5. Young DB, Gideon HP, Wilkinson RJ (2009) Eliminating latent tuberculosis. Trends
Microbiol 17:183188.
6. Berrington WR, Hawn TR (2007) Mycobacterium tuberculosis, macrophages, and the
innate immune response: Does common variation matter? Immunol Rev 219:167186.
7. Casanova JL, Abel L (2002) Genetic dissection of immunity to mycobacteria: The human model. Annu Rev Immunol 20:581620.
8. Comstock GW (1978) Tuberculosis in twins: A re-analysis of the Prophit survey. Am Rev
Respir Dis 117:621624.
9. Casanova JL, Abel L (2007) Primary immunodeciencies: A eld in its infancy. Science
317:617619.
10. Dorman SE, et al. (2004) Clinical features of dominant and recessive interferon
gamma receptor 1 deciencies. Lancet 364:21132121.
11. Altare F, et al. (1998) Impairment of mycobacterial immunity in human interleukin-12
receptor deciency. Science 280:14321435.
12. Altare F, et al. (1998) Inherited interleukin 12 deciency in a child with bacille
Calmette-Gurin and Salmonella enteritidis disseminated infection. J Clin Invest 102:
20352040.
13. Jouanguy E, et al. (1996) Interferon-gamma-receptor deciency in an infant with fatal
bacille Calmette-Gurin infection. N Engl J Med 335:19561961.
14. Newport MJ, et al. (1996) A mutation in the interferon-gamma-receptor gene and
susceptibility to mycobacterial infection. N Engl J Med 335:19411949.
15. de Bakker PI, Telenti A (2010) Infectious diseases not immune to genome-wide association. Nat Genet 42:731732.
16. Thye T, et al.; African TB Genetics ConsortiumWellcome Trust Case Control Consortium (2010) Genome-wide association analyses identies a susceptibility locus for
tuberculosis on chromosome 18q11.2. Nat Genet 42:739741.
17. Cooper AM (2009) Cell-mediated immune responses in tuberculosis. Annu Rev Immunol 27:393422.
18. Smyth GK (2004) Linear models and empirical Bayes methods for assessing differential
expression in microarray experiments. Stat Appl Genet Mol Biol 3:Article3.
19. Huang Q, et al. (2001) The plasticity of dendritic cell responses to pathogens and their
components. Science 294:870875.
20. Tailleux L, et al. (2003) Constrained intracellular survival of Mycobacterium tuberculosis in human dendritic cells. J Immunol 170:19391948.
21. Banchereau J, et al. (2000) Immunobiology of dendritic cells. Annu Rev Immunol 18:
767811.
22. Pickrell JK, et al. (2010) Understanding mechanisms underlying human gene expression variation with RNA sequencing. Nature 464:768772.
23. Stranger BE, et al. (2007) Population genomics of human gene expression. Nat Genet
39:12171224.
24. Berry MP, et al. (2010) An interferon-inducible neutrophil-driven blood transcriptional signature in human tuberculosis. Nature 466:973977.
25. Flynn JL, Chan J (2001) Immunology of tuberculosis. Annu Rev Immunol 19:93129.
26. Korbel DS, Schneider BE, Schaible UE (2008) Innate immunity in tuberculosis: Myths
and truth. Microbes Infect 10:9951004.

Barreiro et al.

DCs were obtained by hybridizing the RNA to the Illumina HumanHT-12 v4

Expression BeadChips arrays. Cytokine and chemokine levels were measured
using the Bio-Plex Pro Human Cytokine 27-plex, according to the manufacturers recommendations. To identify genes whose expression levels were
altered following MTB infection, we used the linear modeling-based approach implemented in the Bioconductor limma package (18). We used
GeneTrail (49) to test for enrichment of functional annotations among differentially expressed genes after MTB infection. Association between SNP
genotypes and either transcript or protein expression levels were examined
by using a linear regression model where each phenotype was regressed
against genotype. In all cases, we assumed that SNPs affecting the phenotype
did so in an additive manner. We mapped infected and noninfected DCs
separately. All regressions were performed using a Python script, whereas
downstream analyses were carried out using the R statistical framework.
ACKNOWLEDGMENTS. We thank M. Stephens for helpful discussions;
Z. Gauhar, J. Tung, D. Cusanovich, L. Quintana-Murci, B. Jabri, J. Maranville,
and F. Luca for comments on the manuscript; G. Stewart for a gift of the
MTB strain used in this study; and T. Thye for sharing the GWAS data with us.
This study was funded by National Institutes of Health Grant AI087658 (to
Y.G. and L.T.). L.B.B. was supported by an Human Frontier Science Program
postdoctoral fellowship. A.A.P. was supported by an American Heart
Association fellowship.
27. Brinckmann J, et al. (2010) Enhanced brillin-2 expression is a general feature of
wound healing and sclerosis: Potential alteration of cell attachment and storage of
TGF-beta. Lab Invest 90:739752.
28. Ramirez F, Rifkin DB (2009) Extracellular microbrils: Contextual platforms for
TGFbeta and BMP signaling. Curr Opin Cell Biol 21:616622.
29. Turner M, et al. (1991) Induction of the interleukin 1 receptor antagonist protein by
transforming growth factor-beta. Eur J Immunol 21:16351639.
30. Muzio M, Sironi M, Polentarutti N, Mantovani A, Colotta F (1994) Induction by
transforming growth factor-beta 1 of the interleukin-1 receptor antagonist and of its
intracellular form in human polymorphonuclear cells. Eur J Immunol 24:31943198.
31. Bellamy R, et al. (1998) Assessment of the interleukin 1 gene cluster and other candidate gene polymorphisms in host susceptibility to tuberculosis. Tuber Lung Dis 79:8389.
32. Wilkinson RJ, et al. (1999) Inuence of polymorphism in the genes for the interleukin
(IL)-1 receptor antagonist and IL-1beta on tuberculosis. J Exp Med 189:18631874.
33. Maranville JC, et al. (2011) Interactions between glucocorticoid treatment and cisregulatory polymorphisms contribute to cellular response phenotypes. PLoS Genet 7:
e1002162.
34. Smirnov DA, Morley M, Shin E, Spielman RS, Cheung VG (2009) Genetic analysis of
radiation-induced changes in human gene expression. Nature 459:587591.
35. Schreiber HA, Sandor M (2010) The role of dendritic cells in mycobacterium-induced
granulomas. Immunol Lett 130:2631.
36. Jiao X, et al. (2002) Dendritic cells are host cells for mycobacteria in vivo that trigger
innate and acquired immunity. J Immunol 168:12941301.
37. Tian T, Woodworth J, Skld M, Behar SM (2005) In vivo depletion of CD11c+ cells
delays the CD4+ T cell response to Mycobacterium tuberculosis and exacerbates the
outcome of infection. J Immunol 175:32683272.
38. Wolf AJ, et al. (2007) Mycobacterium tuberculosis infects dendritic cells with high
frequency and impairs their function in vivo. J Immunol 179:25092519.
39. Bigley V, et al. (2011) The human syndrome of dendritic cell, monocyte, B and NK
lymphoid deciency. J Exp Med 208:227234.
40. Hambleton S, et al. (2011) IRF8 mutations and human dendritic-cell immunodeciency. N Engl J Med 365:127138.
41. Sturgill-Koszycki S, et al. (1994) Lack of acidication in Mycobacterium phagosomes
produced by exclusion of the vesicular proton-ATPase. Science 263:678681.
42. Armstrong JA, Hart PD (1975) Phagosome-lysosome interactions in cultured macrophages infected with virulent tubercle bacilli. Reversal of the usual nonfusion pattern
and observations on bacterial survival. J Exp Med 142:116.
43. Salojin K, Oravecz T (2007) Regulation of innate immunity by MAPK dual-specicity
phosphatases: Knockout models reveal new tricks of old genes. J Leukoc Biol 81:860869.
44. Brooks MN, et al. (2011) NOD2 controls the nature of the inammatory response and
subsequent fate of Mycobacterium tuberculosis and M. bovis BCG in human macrophages. Cell Microbiol 13:402418.
45. Divangahi M, et al. (2008) NOD2-decient mice have impaired resistance to Mycobacterium tuberculosis infection through defective innate and adaptive immunity.
J Immunol 181:71577165.
46. Pandey AK, et al. (2009) NOD2, RIP2 and IRF5 play a critical role in the type I interferon response to Mycobacterium tuberculosis. PLoS Pathog 5:e1000500.
47. Smith AN, et al. (2003) Revised nomenclature for mammalian vacuolar-type H+ -ATPase subunit genes. Mol Cell 12:801803.
48. Xu S, et al. (1994) Intracellular trafcking in Mycobacterium tuberculosis and Mycobacterium avium-infected macrophages. J Immunol 153:25682578.
49. Backes C, et al. (2007) GeneTrailadvanced gene set enrichment analysis. Nucleic
Acids Res 35:W186192.

PNAS | January 24, 2012 | vol. 109 | no. 4 | 1209

GENETICS

response eQTL associated with other infectious diseases, and

that such work will allow us to further classify specic MTBresponse eQTL on the one hand and genetic variation that is
associated with more general variation in the response to infectious agents on the other hand.